It bedevils other carbon pricing schemes as well. It is how to protect emissions-intensive but trade-exposed (EITE)sectors from potentially lethal competition — whether that is in export or domestic markets — from international competitors which do not face a carbon price.
The risk, in the jargon, is "leakage". That happens when carbon pricing results in production moving to countries with higher emissions intensity but lax environmental standards.
The country imposing the carbon price loses out economically and the planetary commons, the climate system, is worse off too.
For example, globally, more than half the emissions arising from aluminium production come not from the chemistry of the smelting process itself, but from generating the electricity needed to power it.
By contrast, electricity-related emissions for the Tiwai Point smelter are only about 10 per cent of the total. If it were to close and the demand for the metal it produces was met from smelters in, say, China, then New Zealand would lose a $1 billion exporter and hundreds of jobs — and global emissions would be higher.
The emissions trading scheme (ETS) deals with this problem by the free allocation of units to EITE firms, which has the effect of limiting their exposure to the carbon price to 10 per cent of their emissions (or 30 per cent for the less exposed).
Last year 6.7 million such units were doled out, notionally worth around $160 million at current ETS prices and covering about 8 per cent of the country's gross emissions of greenhouse gases.
The main recipients were NZ Aluminium Smelters, NZ Steel, Methanex and the pulp and paper sector.
The scope of the free allocation is set to reduce, by 1 per cent a year through the 2020s, then 2 per cent a year through the 2030s and 3 per cent a year in the 2040s. Combined with expected rises in the carbon price, that will continue to exert pressure on EITE firms to reduce emissions if they can.
A report on the issue by economic consultancy Sense Partners last year is sceptical about whether this will save them: "[H]istorically weak innovation and comparatively poor productivity growth are reasons to doubt whether innovation and adaptation by New Zealand firms will be sufficient to overcome potentially wide cost differentials."
How wide those cost differentials will be depends crucially on how climate policy evolves among New Zealand's trading partners.
Before turning to that, it should be noted that the leakage issue is central to the debate over whether to include the agricultural gases, methane and nitrous oxide, in the ETS.
If they are included, the Labour-New Zealand First coalition agreement includes a provision that there be free allocation for 95 per cent of emissions. That is, the ETS price would only apply to 5 per cent of pastoral farming emissions, matching the original degree of protection for the smokestack emitters.
"Leakage risk in agriculture is more subtle than in some other sectors," the Sense Partners report says.
"A good deal of any cost impact from climate policy can be expected to be absorbed by the cost of land — with costs borne by current owners who will face a reduction in asset value.
Land may also move out of more emissions-intensive livestock production and into other forms of production such as horticulture or forestry."
More generally, threatened competitiveness and the risk of leakage are a problem of carbon pricing schemes everywhere.
The theoretical remedy is to bolster it with some form of border carbon adjustment — a carbon tariff on imports, perhaps combined with some form of rebate or tax relief for carbon-intensive exports — to level the playing field.
Border carbon adjustments are a standard feature of proposals for a carbon tax in the United States, for example that of current Democratic Party front-runner Joe Biden.
And not only politicians. This year the Wall Street Journal published an "Economists' Statement on Carbon Dividends" signed by more than 3500 US economists, including 27 Nobel prizewinners and all four living former chairs of the Federal Reserve, going back to Paul Volcker.
They called for the US to adopt a carbon tax, adding that "To prevent carbon leakage and protect US competitiveness a border carbon adjustment system should be established." It would create an incentive for other nations to adopt similar carbon pricing, they said.
It is remarkable to have that many economic grandees agree on anything, let alone something what would impede free trade.
In Europe, a similar measure is supported by French President Emmanuel Macron and the incoming president of the European Commission, Ursula von der Leyen.
The idea is that the way to address what is essentially a free rider problem is to have a common, harmonised global price on carbon.
And the best way to achieve that is for large chunks of the global economy, like the US and European Union, to say, "If you want to export to us, you have to have a comparable carbon price to what we impose on our emitters, or we will ping you at the border to even things out."
As a recent paper for US think tank the Tax Policy Center puts it: "One can imagine negotiations in which the United States (having adopted a carbon tax) grants countries it deems sufficiently climate ambitious an exemption from its BCAs ... In this way the threat or impact of border adjustments, assuming they survive WTO adjudications, could pressure everyone to price carbon — or at least the carbon in their exported products."
Adair Turner, a former chairman of the UK Committee on Climate Change, sees carbon tariffs as an idea whose time has come, driving a "race to the top" in which roughly equal carbon prices spread around the world.
But when the UK committee proposed one 10 years ago it was met with a wall of opposition, with claims it would violate World Trade Organisation rules and unleash tit-for-tat tariff increases justified by whatever environmental priority each country wished to pursue.
The TPC paper makes it clear that the task of calculating tariff rates based on foreign pollution levels would be a nightmare of complexity, even if it were allowed under international trade agreements.
It is hard enough to secure agreement even within the European Union on the stringency and distribution of climate policy. Introducing a controversial and divisive measure like border carbon adjustments into the already fraught multilateral climate process could backfire massively.
So even in a period when multilateralism is in retreat and the big boys are more ready that usual to throw their weight around, it would be unwise to assume that the leakage problem for the New Zealand ETS is going away any time soon.
That said, there may be a case for moving away from the current one-size-fits-all approach to determining eligibility for free allocation.
As the Sense Partners report points out, for the handful of companies involved, carbon price is not the key determinant of commercial viability. Their profitability is affected by prices for a number of key inputs as well as their products.
A report on their behalf from the Castalia consultancy says "EITE businesses welcome regular dialogue with government on effective emissions reduction policies that consider the unique nature of each EITE industry."
There is a precedent for a more bespoke approach. Back when the previous Labour Government planned a carbon tax, it completed a couple of negotiated greenhouse agreements on the terms of exemption from it, including one for the Marsden Point oil refinery which is still in effect.